Homer Nazeran

Automatic sleep stage classification based on EEG signals by using neural networks and wavelet packet coefficients

Currently in the world there is an alarming number of people who suffer from sleep disorders. A number of biomedical signals, such as EEG, EMG, ECG and EOG are used in sleep labs among others for diagnosis and treatment of sleep related disorders. The usual method for sleep stage classification is visual inspection by a sleep specialist. This is a very time consuming and laborious exercise. Automatic sleep stage classification can facilitate this process. The definition of sleep stages and the sleep literature show that EEG signals are similar in Stage1 of non-rapid eye movement (NREM) sleep and rapid eye movement (REM) sleep. Therefore, in this work an attempt was made to classify four sleep stages consisting of Awake, Stage1 + REM, Stage 2 and Slow Wave Stage based on the EEG signal alone. Wavelet packet coefficients and artificial neural networks were deployed for this purpose. Seven all night recordings from Physionet database were used in the study. The results demonstrated that these four sleep stages could be automatically discriminated from each other with a specificity of 94.4 ± 4.5%, a of sensitivity 84.2+3.9% and an accuracy of 93.0 ± 4.0%.

Reducing power line interference in digitised electromyogram recordings by spectrum interpolation.

Interference from power lines (50 or 60 Hz) is the largest source of extraneous noise in many bio-electric signals and is within the bandwidth of many such signals. In this study, two different methods were compared for their efficacy in removing 50 Hz noise added to surface electromyogram (EMG) signals free of power line interference. The first was a simple second-order recursive digital notch filter. The second was an approach called spectrum interpolation, in which it is assumed that the magnitude of the original 50 Hz component of the EMG signal can be approximated by interpolation of the amplitude spectrum of the signal. When the spectrum was based on records containing an integer number of cycles of 50 Hz interference, and the frequency resolution was finer than 1 Hz, spectrum interpolation performed similarly to, or significantly better than, the notch filter (p<0.01). It was also possible to make spectrum interpolation more robust than the notch filter. The Pearson squared correlation coefficient r2 between clean signals and signals processed using the notch filter was reduced from 0.98 to 0.65 when the interference frequency was increased by 0.5 Hz, but r2 for spectrum interpolation at 0.2 Hz resolution was only reduced from 0.99 to 0.85 if spectral values between approximately 49.5 and 50.5 Hz were modified by interpolation.

Automatic sleep staging using empirical mode decomposition, discrete wavelet transform, time-domain, and nonlinear dynamics features of heart rate variability signals

The conventional method for sleep staging is to analyze polysomnograms (PSGs) recorded in a sleep lab. The electroencephalogram (EEG) is one of the most important signals in PSGs but recording and analysis of this signal presents a number of technical challenges, especially at home. Instead, electrocardiograms (ECGs) are much easier to record and may offer an attractive alternative for home sleep monitoring. The heart rate variability (HRV) signal proves suitable for automatic sleep staging. Thirty PSGs from the Sleep Heart Health Study (SHHS) database were used. Three feature sets were extracted from 5- and 0.5-min HRV segments: time-domain features, nonlinear-dynamics features and time-frequency features. The latter was achieved by using empirical mode decomposition (EMD) and discrete wavelet transform (DWT) methods. Normalized energies in important frequency bands of HRV signals were computed using time-frequency methods. ANOVA and t-test were used for statistical evaluations. Automatic sleep staging was based on HRV signal features. The ANOVA followed by a post hoc Bonferroni was used for individual feature assessment. Most features were beneficial for sleep staging. A t-test was used to compare the means of extracted features in 5- and 0.5-min HRV segments. The results showed that the extracted features means were statistically similar for a small number of features. A separability measure showed that time-frequency features, especially EMD features, had larger separation than others. There was not a sizable difference in separability of linear features between 5- and 0.5-min HRV segments but separability of nonlinear features, especially EMD features, decreased in 0.5-min HRV segments. HRV signal features were classified by linear discriminant (LD) and quadratic discriminant (QD) methods. Classification results based on features from 5-min segments surpassed those obtained from 0.5-min segments. The best result was obtained from features using 5-min HRV segments classified by the LD classifier. A combination of linear/nonlinear features from HRV signals is effective in automatic sleep staging. Moreover, time-frequency features are more informative than others. In addition, a separability measure and classification results showed that HRV signal features, especially nonlinear features, extracted from 5-min segments are more discriminative than those from 0.5-min segments in automatic sleep staging.

Digital phonocardiography: a PDA-based approach

In this paper we demonstrate the applicability of advanced digital signal processing algorithms to the analysis of heart sound signals and describe the development of a PDA-based biomedical instrument capable of acquisition, processing, and analysis of heart sounds. Fourier transform-based spectral analysis of heart sounds was carried out first to show the differences in the frequency contents of normal and abnormal heart sounds. As the time-varying nature of heart sounds calls for better techniques capable of analyzing such signals: the short time Fourier transform (STFT) or spectrogram analysis was performed next. This method performed remarkably well in displaying frequency, magnitude, and time information of the heart sounds, providing robust parameters to make accurate diagnosis. With continuous technological advancements in computing and biomedical instrumentation, and the concurrent popularity of handheld instruments in the medical community, we introduce the concept of PDA-based digital phonocardiography. A prototype system is comprised of a digital stethoscope and a pocket PC. Heart sounds are recorded and displayed in the pocket PC screen. Advanced signal processing algorithms are implemented using the combined capabilities of software tools such as LabVlEW and embedded Visual C/sup ++/.

Non-Invasive Fetal Monitoring: A Maternal Surface ECG Electrode Placement-Based Novel Approach for Optimization of Adaptive Filter Control Parameters Using the LMS and RLS Algorithms

This paper is focused on the design, implementation and verification of a novel method for the optimization of the control parameters (such as step size μ and filter order N) of LMS and RLS adaptive filters used for noninvasive fetal monitoring. The optimization algorithm is driven by considering the ECG electrode positions on the maternal body surface in improving the performance of these adaptive filters. The main criterion for optimal parameter selection was the Signal-to-Noise Ratio (SNR). We conducted experiments using signals supplied by the latest version of our LabVIEW-Based Multi-Channel Non-Invasive Abdominal Maternal-Fetal Electrocardiogram Signal Generator, which provides the flexibility and capability of modeling the principal distribution of maternal/fetal ECGs in the human body. Our novel algorithm enabled us to find the optimal settings of the adaptive filters based on maternal surface ECG electrode placements. The experimental results further confirmed the theoretical assumption that the optimal settings of these adaptive filters are dependent on the ECG electrode positions on the maternal body, and therefore, we were able to achieve far better results than without the use of optimization. These improvements in turn could lead to a more accurate detection of fetal hypoxia. Consequently, our approach could offer the potential to be used in clinical practice to establish recommendations for standard electrode placement and find the optimal adaptive filter settings for extracting high quality fetal ECG signals for further processing. Ultimately, diagnostic-grade fetal ECG signals would ensure the reliable detection of fetal hypoxia.

A Phonocardiographic-Based Fiber-Optic Sensor and Adaptive Filtering System for Noninvasive Continuous Fetal Heart Rate Monitoring

This paper focuses on the design, realization, and verification of a novel phonocardiographic- based fiber-optic sensor and adaptive signal processing system for noninvasive continuous fetal heart rate (fHR) monitoring. Our proposed system utilizes two Mach-Zehnder interferometeric sensors. Based on the analysis of real measurement data, we developed a simplified dynamic model for the generation and distribution of heart sounds throughout the human body. Building on this signal model, we then designed, implemented, and verified our adaptive signal processing system by implementing two stochastic gradient-based algorithms: the Least Mean Square Algorithm (LMS), and the Normalized Least Mean Square (NLMS) Algorithm. With this system we were able to extract the fHR information from high quality fetal phonocardiograms (fPCGs), filtered from abdominal maternal phonocardiograms (mPCGs) by performing fPCG signal peak detection. Common signal processing methods such as linear filtering, signal subtraction, and others could not be used for this purpose as fPCG and mPCG signals share overlapping frequency spectra. The performance of the adaptive system was evaluated by using both qualitative (gynecological studies) and quantitative measures such as: Signal-to-Noise Ratio—SNR, Root Mean Square Error—RMSE, Sensitivity—S+, and Positive Predictive Value—PPV.

ECG noise cancellation using digital filters

A digital filter structure is proposed to maximally remove noise from the ECG signals. This structure is based on cascading a zero-phase bandpass, an adaptive filter, and multi-band-pass filter. It provides an efficient method for removing noise from the ECG signals. This filter structure has low implementation complexity and introduces little noise into a typical ECG. It can be applied to real-time applications particularly automatic cardiac arrhythmia classifiers.

Wireless photoplethysmographic device for heart rate variability signal acquisition and analysis

The photoplethysmographic (PPG) signal has the potential to aid in the acquisition and analysis of heart rate variability (HRV) signal: a non-invasive quantitative marker of the autonomic nervous system that could be used to assess cardiac health and other physiologic conditions. A low-power wireless PPG device was custom-developed to monitor, acquire and analyze the arterial pulse in the finger. The system consisted of an optical sensor to detect arterial pulse as variations in reflected light intensity, signal conditioning circuitry to process the reflected light signal, a microcontroller to control PPG signal acquisition, digitization and wireless transmission, a receiver to collect the transmitted digital data and convert them back to their analog representations. A personal computer was used to further process the captured PPG signals and display them. A MATLAB program was then developed to capture the PPG data, detect the RR peaks, perform spectral analysis of the PPG data, and extract the HRV signal. A user-friendly graphical user interface (GUI) was developed in LabView to display the PPG data and their spectra. The performance of each module (sensing unit, signal conditioning, wireless transmission/reception units, and graphical user interface) was assessed individually and the device was then tested as a whole. Consequently, PPG data were obtained from five healthy individuals to test the utility of the wireless system. The device was able to reliably acquire the PPG signals from the volunteers. To validate the accuracy of the MATLAB codes, RR peak information from each subject was fed into Kubios software as a text file. Kubios was able to generate a report sheet with the time domain and frequency domain parameters of the acquired data. These features were then compared against those calculated by MATLAB. The preliminary results demonstrate that the prototype wireless device could be used to perform HRV signal acquisition and analysis.

Wavelet-based EEG denoising for automatic sleep stage classification

In automatic sleep stage classification, as in any other signal processing task involving the easily contaminated EEG signals, denoising constitutes a crucial pre-processing step that must be addressed before carrying out further analysis on the EEG signals. Discrete wavelet transform offers an effective solution for denoising nonstationary signals such as EEG due to its shrinkage property. In this paper, we explored the application of wavelet denoising method to EEG signals acquired during different sleep stages classified according to the RK rules, with the objective to identify suitable thresholding rules and threshold values. Preliminary results showed that the combination of soft thresholding rule applied to the Detailed wavelet coefficients with the Universal threshold value produced better performance measures such as a smaller Minimum Squared Error (MSE) and a larger signal-to-Noise Ratio (SNR). Similarly improved results were obtained for Stage 1, Stage 2, Stage 3, Stage 4 and REM stage EEG signals using this combination. Such thresholding rule and values are equally well applicable to denoising EEG epochs acquired from deep sleep stages.

A Non-Invasive Multichannel Hybrid Fiber-Optic Sensor System for Vital Sign Monitoring

In this article, we briefly describe the design, construction, and functional verification of a hybrid multichannel fiber-optic sensor system for basic vital sign monitoring. This sensor uses a novel non-invasive measurement probe based on the fiber Bragg grating (FBG). The probe is composed of two FBGs encapsulated inside a polydimethylsiloxane polymer (PDMS). The PDMS is non-reactive to human skin and resistant to electromagnetic waves, UV absorption, and radiation. We emphasize the construction of the probe to be specifically used for basic vital sign monitoring such as body temperature, respiratory rate and heart rate. The proposed sensor system can continuously process incoming signals from up to 128 individuals. We first present the overall design of this novel multichannel sensor and then elaborate on how it has the potential to simplify vital sign monitoring and consequently improve the comfort level of patients in long-term health care facilities, hospitals and clinics. The reference ECG signal was acquired with the use of standard gel electrodes fixed to the monitored person’s chest using a real-time monitoring system for ECG signals with virtual instrumentation. The outcomes of these experiments have unambiguously proved the functionality of the sensor system and will be used to inform our future research in this fast developing and emerging field.